B.Tech student in Computer Science Engineering (AI & DS) at SASTRA Deemed University, Thanjavur — based in Hosur, India. I sit at the intersection of reinforcement learning, deep learning, and quantitative finance, currently building DNN‑RL hybrid models that forecast exchange rates using macroeconomic signals: crude oil, gold futures, and the NIFTY 50.
Alongside research, I build full-stack AI applications, compete in national hackathons, and dig into network security and multi-agent systems whenever a good problem shows up.
|
🎓 Education
|
📍 Currently
|
|
Mar 2026 – Present · SASTRA Deemed University DNN‑RL hybrid model combining deep neural networks with PPO policy-gradient reinforcement learning to augment time-series forecasting. Incorporates crude oil, gold futures, and NIFTY 50 as macroeconomic signals.
Result: 95%+ predictive accuracy |
Feb 2026 – Present · SASTRA Deemed University Investigating Advanced Persistent Threat detection and anomaly identification with ML, aimed at enterprise-grade network security frameworks against long-term, sophisticated threat actors.
Status: active research |
| 🧠 AI / RL / Deep Learning |
|
| 💬 NLP / LLM |
|
| 👁️ Computer Vision |
|
| 💻 Languages |
|
| 🌐 Web / Full Stack |
|
| 📊 Data Science |
|
| 📈 Quant Finance |
|
| 🔧 Tools / Systems |
|
BRAIN Research Consultant · WorldQuant — May 2026 – Present Developing quantitative alphas on WorldQuant's BRAIN platform, applying probability theory and statistical modeling to construct and backtest financial signal strategies across global markets.
Freelance AI/ML Developer · Freelancer.in — 2026 – Present Delivering custom AI/ML solutions and full-stack apps; building end-to-end data pipelines, training workflows, and deployment-ready APIs on the Django/Python stack.
Student Researcher · SASTRA Deemed University — Feb 2026 – Present DNN-RL exchange rate forecasting and APT detection research, both tracked toward publication.
| Project | Description | Stack |
|---|---|---|
| RL-Based Traffic Management (SIH 2025) | Multi-agent RL traffic-signal control minimizing clearance time across intersections; OpenCV on Raspberry Pi for real-time vehicle detection | DQN MARL OpenCV Raspberry Pi |
| Exchange Rate Forecasting Engine | DNN + PPO reinforcement learning agents on multivariate financial time series (crude oil, gold, index data) — 95%+ accuracy | TensorFlow PPO Time Series |
| Web-Based AI Applications | Full-stack platforms with AI/ML-backed features, end-to-end from data pipeline to UI | Django MongoDB JavaScript |
🥇 CMI STEMS Top 30 India — Chennai Mathematical Institute, research aptitude · 2025 📝 IEEE Research Paper (in progress) — DNN-RL exchange rate prediction · 2026 target 🚦 Smart India Hackathon 2025 — RL-based intelligent traffic management, Odisha 🏛️ NPTEL SWAYAM — Deep Learning: architectures, backprop & optimization · 2026 📊 Yale (Coursera) — Financial Markets, market theory & quant finance 🌐 Harvard CS50W — Web Programming with Python & JavaScript · in progress ➗ Mathematics Olympiad — IOQM qualifier, merit certificate holder
Open to research collaborations, internships, freelance work, and conversations about AI, quant finance, or anything at the edge of both.
mithilcuber@gmail.com · LinkedIn · GitHub · Freelancer
© 2026 Mithilesh Adhinarayanan · Hosur, India · built with a bit too much PPO